Solution dynamics of tris(acetylacetonato)chromium(III) - Inorganic

V.M. Nekipelov , K.I. Zamaraev. Coordination Chemistry Reviews 1985 61, ... Nobuo Suzuki , Mutsuko Itoh , Hitoshi Watarai. Polyhedron 1982 1 (4), 383-...
0 downloads 0 Views 414KB Size
Inorganic Chemistry, Vol. 16, No. 3, 1977 709

Notes (11) J. S. Griffith, “The Theory of Transition Metal Ions”, Cambridge University Press, Cambridge, 1971. (12) J. T. Wrobleski and G. J. Long, Appl. Spectrosc., in press. (13) F. E. Mabbs and D. J. Machin, “Magnetism and Transition Metal Complexes”, Chapman and Hall, London, 1973, p 5 . (14) F. Bohlmann, Chem. Ber., 91, 2157 (1958). (15) M. Wiewiorowski and J. Skolik, Bull. Acad. Pol. Sci., Ser. Sci. Chim., 10, 1 (1962). (16) M. Wiewiorowski, 0. E. Edwards, and M. D. Bratek-Wiewiorowski,Can. J . Chem., 45, 1447 (1967). (1 7) J. T. Wrobleski, B. F. Little, and V. K. Hardman, Abstracts of Papers, 10th Midwest Regional Meeting of the American Chemical Society, 1974, No. 420, p 66. (18) K. Nakamoto. “Infrared Soectra of Inoraanic and Coordination Compounds”, 2nd ed,Wiley-Iiterscience, New York, N.Y., 1970, p 187 ff. (19) Y. S.Wong, S. Jacobson, P. C. Chieh, and A. J. Carty, Inorg. Chem., 13, 284 (1974). (20) A. J. Carty and A. Efraty, Can. J . Chem., 47, 2573 (1969). (21) G. L. Kulasingam and W. R. McWhinnie, J. Chem. Sot. A, 254 (1968). (22) J. L. Burmeister, Coord. Chem. Reu., 1, 205 (1966). (23) K. Nakamoto, Angew. Chem., Int. Ed. Engl., 11, 666 (1972).

(24) D. M. Adams, “Metal-Ligand and Related Vibrations”, St. Martin’s Press, New York, N.Y., 1968. (25) Reference 12, p 107 ff. (26) Supplementary material. (27) F. A. Cotton, D. M. L. Goodgame, and M. Goodgame, J . Am. Chem. SOC.,83,4690 (1961); F. A. Cotton, D. M. L. Goodgame, and A. Sacco, ibid., 83,4157 (1961); F. A. Cotton and M. Goodgame, ibid., 83, 1777 (1961). (28) C. Simo and S. Holt, Inorg. Chem., 7, 2655 (1968); R. L. Carlin and S. L. Holt, Jr., ibid., 2, 849 (1963); J. Ferguson, J . Chem. Phys., 32, 528 (1960). (29) R. C. Dickinson and G. J. Long, Inorg. Chem., 13, 262 (1974). (30) G. J.,,Long and D. L. Coffen, Inorg. Chem., 13, 270 (1974). (31) E. Konig, Struct. Bonding (Berlin), 9, 175 (1971). (32) A. B. P. Lever and S. M. Nelson, J. Chem. Sot. A, 859 (1966). (33) C. K. Jorgensen, “Absorption Spectra and Chemical Bonding in Complexes”, Pergamon Press, Oxford, 1962, p 107 ff. (34) A. Flamini, L. Sestili, and C . Furlani, Inorg. Chim.Acta, 5, 241 (1971). (35) E. Menzel, W. R. Vincent, D. K. Johnson, G. L. Seebach, and J. R. Wasson, Inorg. Chem., 13, 2465 (1974). (36) M. Gerlock and R. C. Slade, “Ligand-Field Parameters”, Cambridge University Press, Cambridge, 1973.

Notes Contribution from the Department of Chemistry, University of Alabama in Birmingham, Birmingham, Alabama 35294

Solution Dynamics of Tris(acetylacetonato)chromium(III) Gerald S. Vigee and Charles L. Watkins’

Received May 28, I976

AIC60401X

Studies of the solution dynamics of transition metal complexes are important in understanding the catalytic activity of metal ions in both biological and nonbiological systems.’ In particular, we are interested in the chemical dynamics of first and second coordination sphere exchange of ligands with bulk solvent molecules. Such studies are capable of exploring the rate, order, and preferential orientation of exchanging species in the first and second coordination sphere and relating these parameters to intrinsic properties such as ion size and charge, steric factors of association between the metal ion complex and solvent molecules, and dipole moment and hydrogen-bonding ability of the ~ o l v e n t . ~ - ~ Certain paramagnetic complexes such as tris(acety1acetonato)chromium(III), C r ( a ~ a c )have ~ , been used extensively as spin relaxants in carbon-13 Fourier-transform NMR studies. The controversy and questions which have arisen over their use are intimately associated with the first and second coordination sphere behavior of the solvent and solute molecules about the paramagnetic It has been postulated that while the spin-lattice and spin-spin relaxation times of the carbon- 13 species are reduced in the presence of Cr(II1) complexes and the NOE (nuclear Overhauser effect) is quenched, the chemical shift is unaffected. Our previous studies of the solution dynamics of tris(ethy1enediamine)chromium(II1) and hexakis(dimethy1 sulfoxide)chromium(III) indicate that this assumption may not be t r ~ e . ’ ~ J ~ Here we report the dependence of the chemical dynamics of first and second coordination sphere exchange of C r ( a ~ a c ) ~ with four solvents on various properties of the solvents such as dipole moment, basicity, and hydrogen-bonding ability. In addition to its role as a spin relaxant, C r ( a ~ a c was ) ~ chosen for the present study because of its well-known kinetic and thermodynamic stability and because of the nature of the d3 isotropic electronic ground state which makes the interpretation of the N M R of its solutions less complicated. The changes in carbon- 13 and hydrogen- 1 chemical shifts upon addition

lo



311

313

3;

37

3’9

41

43

103 T

Figure 1. Line broadening of the methyl protons of acetonitrile for 0.1 M Cr(acac)j in acetonitrile.

of the C r ( a ~ a c )are ~ reported for the solvents. Experimental Section Proton line broadening and chemical shift measurements and carbon-13 chemical shift and T , measurements were made on each of the four solvents, benzene, acetonitrile, methanol, and dimethyl sulfoxide (DMSO), in the absence and presence of C r ( a ~ a c ) ~Data . resulting from the temperature dependence of the proton line broadening were analyzed using the methods of Swift and Connick.14 All data were normalized to 0.1 M Cr(acac)3where the concentration of C r ( a ~ a c )ranged ~ from 0.05 to 0.1 M. The proton N M R data were obtained using Varian A-60A and EM-390 spectrometers. The carbon-1 3 data were obtained using a Bruker HX-90 FT spectrometer with a Nicolet 1085 data system. T M S was used as an internal reference for the chemical shift studies. All solvents were of spectrochemical quality. The Cr(acac)3, obtained commercially, was purified by recrystallization.

Results and Discussion Figure 1 shows a Swift-Connick plot for C r ( a ~ a c )in~ acetonitrile. The results for acetonitrile are also typical of the results for benzene and DMSO. The plot is linear with positive slope; A I P for the single process observed is 2.44 kcalfmol. From the temperature dependence of the relaxation rate, Tzp’, and the value of AHt, exchange is fast as defined by the Swift-Connick re1ation~hip.I~ In the fast-exchange region, the Swift-Connick equation reduces to l/Tz, = f / T z m ,where f

710 Inorganic Chemistry, Vol. 16, No. 3, 1977

Notes Table I. Second Coordination Sphere Parameters for Some Chromium Complexes in Dimethyl Sulfoxide

/

4 9

30

34

38

42

46

50

103 T

Figure 2. Line broadening of the methyl and hydroxyl protons of methanol for 0.1 M Cr(acac)3 in methanol.

is a probability factor and 1/T2* is the relaxation rate of the coordinated resonating nucleus. In the fast-exchange region, 1/T2,,, is independent of the mean lifetime of the coordinated ligand. The exchange process is expected to be second coordination sphere because of the well-known ligand stability of Cr(acac)+ The ivf values for exchange for benzene and DMSO are 2.96 and 3.41 kcal/mol, respectively, and are also expected to be second coordination sphere proce~ses.'~ Methanol poses an interesting case since the line broadening of both the methyl and the hydroxyl protons may be monitored as a function of temperature. Figure 2 summarizes the results of the line broadening study. The hydroxyl proton relaxation is more rapid than the methyl proton relaxation. Since the outer-sphere dipole-dipole relaxation depends upon the Cr-H distances as l / r 3 , this evidence indicates that the hydroxyl proton is closer to the chromium ion than the methyl proton. This suggests a preferential alignment of the methanol molecule in the second coordination sphere of the metal ion. Fackler,I6 by IR studies, has shown that methanol does bind by hydrogen bonding to the acetylacetonate. Interestingly, the slopes of hydroxyl and methyl plots are different [AP = 2.5 kcal/mol (OH) vs. 3.16 kcal/mol (CH,)]. If line broadening for both hydroxyl and methyl protons occurred by the same mechanism, Le., dipolar coupling to the Cr3+,the T2p1values at the same temperatures should be different due to the l / r 3dipole relationship. However, the slopes ( A p ' s ) would be the same. Therefore, the hydroxyl proton is experiencing an additional relaxation process. Eaton,17 by contact shift and molecular orbital calculations, showed a significant amount of unpaired spin density to be delocalized onto the Cr(acac)3 oxygen. Thus the modulation of the unpaired spin density transfer between the acetylacetonate oxygen and the hydrogen-bonded hydroxyl proton could account for the additional process observed. This proposal is supported by the fact that the hydroxyl proton is the only solvent proton which possesses a contact shift. The DMSO exchange with Cr(acac), can be adequately explained by invoking an outer-sphere dipolar mechanism

Complex

An, Hz

Cr(DMS0),3+ Cr(en)3+ Cr(acac),

-1.2 X IO5 -3.0 X l o 5

k* (25 "C), AH*, s-' kcal/mol

1.3 x l o 3 1.0 X l o 3

-103

6.9 6.6 3.4

although there is evidence of small contact shifts at high Cr(acac), concentrations. It is interesting to contrast the results of the C r ( a ~ a c study ) ~ with our previous studies of tris(ethylenediamine)chromium(III), Cr(en)33+,and hexakis(dimethy1 sulfoxide)chromium(III), Cr(DMS0)63+,in DMSO (Table I). The Swift-Connick plots for Cr(DMS0)63+and Cr(en)33f in DMSO are complex and can be resolved into both first coordination sphere processes at high temperatures (ivf E 25 kcal/mol) and second coordination sphere processes at lower temperatures, as shown in Table I. An examination of the large A P ' s and hyperfine coupling constants, A,, and the measurable exchange rates, k*, for Cr(DMS0)63+and C r ( e r ~ ) ~indicates ~+ that second coordination sphere bonding of DMSO is more significant than in the case of DMSO in solutions of Cr(acac)3 where small AP and A , values are noted and where only a lower limit to the exchange rate can be ca1c~lated.l~ The differences in the kinetic parameters are directly affected by the charge on the chromium complex and are consistent with similar findings by Frankel' and L a n g f ~ r d . ~ Second Coordination Sphere Structure and Exchange Mechanism. The Swift-Connick relationship reduces to 1/ T2 = f / T z munder fast-exchange conditions. The value of T2,,,- P may be determined as

(inner -sphere) 1671 T12p2g2S(S + l ) p [ m ] N i- 9 y o 3 X lo3 (outer sphere)

-{

,-

(contact shift)

where _ AV - -Any,gS(S

+ 1)

YIl(3kr) and y,,ye, g, S, k , T, A,, and h have their usual meaning.]* In the case of C r ( a ~ a c )the ~ , first term, which represents first coordination sphere exchange, may be neglected. The third term, the contact shift term, is unimportant for DMSO, benzene, and acetonitrile solutions and only slightly influences the methanol results. The second term representing outersphere dipolar relaxation may be considered the most important contribution to TZ,-l. The distance between the Cr3+ ion and the resonating proton, ro, may then be calculated for each solvent. 1 / is~defined ~ as 1/T2, + 1 / +~1 ~/ +~~ / ~T F I . ' ~ l / ~ the ~ ,electronic exchange rate, and 1 / ~the , ligand exchange rate, are negligible compared to 1 / ~the ~ ,rotational correlation rate in these systems. 1/T2e, the electronic transverse relaxation rate, is also small compared to 1/r, for chromium solutions.12 Thus, 7,may be considered to be the rotational correlation time 7,for the solvent molecule in the chromium ion system and is calculated using the StokesEinstein equation. A summary of the calculated distances is given in Table 11. The table also gives a comparison of these distances with distances determined from space-filling models and x-ray crystallographic data taking as a standard distance V

Inorganic Chemistry, Vol. 16, No. 3, 1977 711

Notes Table 11. Summary of Model and Calculated Distances 0 - 0 = 2.0 cm = 1.99 Azo) Molecule Methanol (OH first) DMSO (S=O first) Benzene (flat and cleft) Acetonitrile (N first) (CH, first)

Model dist, A

calcd dist, A

3.85 (OH) 4.80 (CH,) 5.15

3.7 4.7 5.0

Preferential

4.6

5.1

Random

5.28 3.09

3.72

Preferential, favors CH, first

Orientation

Preferential

Cr-0 = 1.99 A = 2.9 cmaZoSeveral orientations for each molecule were assumed. While the ,distances should not be taken as absolute values, the good agreement between the calculated distances and distances assuming a given orientation should be indicative of preferential ordering in the outer solvation sphere. For example, methanol is preferentially ordered due to the molecule being oriented along its dipole2 due to its participation in hydrogen bonding with the acetylacetonate oxygen. The methyl proton distance is very close to that calculated. Agreement for the hydroxyl proton is not as good due to the contribution of the contact term to 1/TZm and to the contribution of the additional relaxation mechanism to I/T2,. Best agreement for DMSO is obtained by assuming preferential orientation with the oxygen atoms pointed into the octahedral face of the C r ( a ~ a c ) ~For . benzene, a random orientation of the flat and cleft positions best fits the data. The acetonitrile molecule, from the NMR calculations, has a mildly preferred orientation with the methyl group pointing into the octahedral face of the C r ( a ~ a c ) ~This . unexpected orientation is probably stabilized by weak hydrogen bonding between the methyl protons and acetylacetonate oxygens. The ,methyl protons become capable of hydrogen bonding as a result of the electron-withdrawing nature of the cyano group. The preferential orientation of the methyl group of acetonitrile is further supported by the carbon-13 chemical shift and TI data. The methyl carbon is shifted 0.3 ppm downfield [0.1 ppm C r ( a ~ a c ) while ~ ] the cyano carbon is unshifted. The corrected Ti's for the methyl and cyano carbons are 0.72 and 0.93 s, respectively. Since T I is proportional to l / r 6 for the electronic-nuclear dipole mechanism," the methyl protons must be closer to the paramagnetic center. Although the proton chemical shifts for each solvent were small upon addition of 0.1 M Cr(acac)3, the carbon-13 chemical shifts were larger and of the magnitude reported previou~ly.~ Each of the carbon-13 chemical shifts, except that of the cyano group of the acetonitrile, moves downfield upon addition of the C r ( a ~ a c ) the ~ , range being 0.1-0.3 ppm. Acknowledgment. We express our thanks to Dr. Jerry Glickson for the use of the Bruker HX-90 spectrometer. The project was partially supported by a University College Faculty Research Grant to C.L.W. Registry No. Cr(acac),, 13681-82-8; Cr(en),,+, 15276-13-8; Cr(DMS0)63+, 19122-76-0.

References and Notes (1) (2) (3) (4) (5)

(6) (7) (8)

(9) (10) (1 1) (12)

G. L. Eichorn, Ed., Inorg. Biochem., 1 and 2 (1973). L. S. Frankel, J . Phys. Chem., 73, 3897 (1969). N. S. Angerman and R. B. Jordan, Inorg. Chem., 8, 65, 1824 (1969). T. R. Stengle and C. H. Langford, J . Phys. Chem., 69, 3299 (1965). P. M. Henrichs and S. Gross, J. Magn. Reson., 17, 399 (1975). 0. A. Gansow, A. R. Burke, and W. D. Vernon, J . Am. Chem. Soc., 94, 2550 (1972). G. C. Levy and J. D. Cargioli, J. Magn. Reson., 10, 231 (1973). J. G. Hexem, U. Edlund, and G. C. Levy, J. Chem. Phys., 64,936 (1976). G. C. Levy and U. Edlund, J . Am. Chem. SOC.,97, 4482 (1975). G. C. Levy and R. A. Komoroski, J. Am. Chem. SOC.,96,678 (1974). F. A. Cotton, D. L. Hunter, and A. J. White, Imrg. Chem., 14,703 (1975). C. L. Watkins, G. S. Vigee, and M. E. Harris, J . Phys. Chem., 77,855 (1973).

G. S. Vigee and P. Ng, J . Inorg. Nucl. Chem., 13, 2477 (1971) T. J. Swift and R. E. Connick, J . Chem. Phys., 37, 307 (1962). Z. Luz and S. Meiboom. J. Chem. Phvs.. 40. 2686 (1964). T. S. Davis and J. P. Fackler, Jr., Inoig. 'Chkm., 5 , 242 (i966). D. R. Eaton, J . Am. Chem. SOC.,87, 3097 (1965). L. L. Rusnak and R. B. Jordan, Inorg. Chem., 10,2686 (1971), 14,988 (1974) -, (19) T. J. Swift in "NMR of Paramagnetic Molecules", G. N. La Mar, W. Dew. Horrocks, Jr., and R. H. Holm, Ed., Academic Press, New York, N.Y., 1973. pp 72-77. (20) J. P. Fackler, Jr., and A. Avdeef, Inorg. Chem., 13, 1864 (1974).

(13) (14) (15) (16) (17) (18)

\ - -

Contribution from the Department of Chemistry, University of British Columbia, Vancouver, Canada, V6T 1W5

Intermediate-Spin ( S = 3/2) Porphyrinatoiron(II1) Complexes David H. Dolphin, John R. Sams,* and Tsang Bik Tsin

Received June 1, 1976

AIC604052

There is much current interest in the synthesis of new iron porphyrin complexes having unusual electronic states or coordinations. The most common types of iron porphyrins are the pentacoordinate high-spin ( S = 5 / 2 ) iron(II1) complexes Fe(P)X (P = porphyrinato dianion, X = anionic group), hexacoordinate low-spin ( S = 'I2) iron(II1) complexes Fe(P)L2X (L = amine base), and hexacoordinate low-spin ( S = 0) iron(I1) complexes Fe(P)L2. Recently, preparations of a few iron porphyrin derivatives containing high-spin ( S = 2) and intermediate-spin ( S = 1) iron(I1) have been reported.'-5 The former are t h o ~ g h tto~ ,be~ pentacoordinate with the iron atom substantially out of the plane of the porphyrinato nitrogen atoms, while the latter are tetracoordinate with a planar FeN4 core.3 The only apparently unknown ground state for either iron(I1) or iron(II1) porphyrins is the unusual intermediate-spin (S = 3/2) iron(II1) state. Although this ground state is not known to occur in natural iron prophyrins, there is evidence for a low-lying S = 3 / 2 state in some heme proteins,6aand a new iron-sulfur cluster with S = 3 / 2 has recently been discovered in nitrogenase.6b Moreover, the preparation and characterization of S = 3 / 2 tetraaza iron(II1) systems has recently been reported.' The criterion for obtaining the S = state for a d5 ion is essentially the same as that for obtaining the S = 1 state for a d6 ion: the dX2~y2 antibonding orbital must lie well above the other d orbitals and be unoccupied, while the remaining four d orbitals must fill by Hund's rules. This requires strong bonding in the equatorial ( x y ) directions and very weak bonding (or none) in the axial direction so that dZ2remains close in energy to the t2gorbitals. It therefore seemed likely that the S = state could be stabilized in Fe(P)X type complexes if X were a sufficiently weak-field anion. Ogoshi et a1.* have reported the preparation of octaethylporphyrinatoiron(II1) perchlorate, Fe(OEP)C104, and three of its monoamine adducts. The magnetic moments of all four complexes at 288 K were intermediate between those expected for high-spin and low-spin iron(II1). These results, augmented by IR and reflectance spectra, were interpreted in terms of a 6A1-2Tz spin equilibrium.8 Although the temperature dependence of yefffor the three adducts appears to support this interpretation, only a single FCLeff value at 288 K was reported for Fe(OEP)ClO,. We have now prepared this complex and its bis(ethano1) solvate by a new route and present clear evidence that in both compounds the iron(II1)